论文部分内容阅读
With the development of mobile intemet,Web services tend to be fragmented and heterogeneous.The demands of users are expressed in multi-dimensionalities.Service recommendation is an efFective way to help users discovering service resources that can make the Web service more convenient and intelligent.
Mashup,which is a kind of light-weight method for integrating service functions,provides users with a way to integrate existing service resources to meet their needs.As a user-generated content that integrates service resources,mashups show the collaboration relationship between service resources.In the service resource community,this kind of relationship between service resources embodies their functional connection.indicating their functional substitutability and compatibility.Besides,the collaboration network shows the knowledge process within the community,which suggests a way of recommending service according to their functional similarity.However,currently,how to extract this service collaboration network and combine them with service recommendations is still an unsolved question.
The service recommendation method based on context-aware is also a critical way of optimizing the recommendation performance.In response to the complex internet environment,context-aware techniques can mine structured demands expressions from the users situation information and calculate the recommendation results accordingly.
The goal of this study is to fill four gaps:(l)Extract the resources features from service collaboration network metrics and construct a service recommender;(2)Optimize the network projection weights to balance the influence of popular services;(3)Use MLP to extract user demands from textual information;(4)Match the usersdemands with service resources via service collaboration network.
The main work of this study is constructing a service recommendation framework.including two major modules:Recommendation computation based on the service collaboration network extraction and context extraction.The recommendation computation uses network matrix multiplication and resource-allocation process to transform the network into service features.Then the module uses the cosine method to compute the similarity between application functions and services.The context extraction module transforms the text into structured features by using TF-IDF and TSVD.Then the features are used to train a multilayers perceptron,which can learn application context from input features.This module also applies a cross-validation model selector to train the supervised model.
The construction process of the two modules constitutes the offline data processingsubsystem.The users input their demands in natural language via the user interface.and the online recommendation subsystem will use the trained models and recommendation scores to decide the recommendation results.
In the empirical part,this study uses the data from Programmable Web to implement this service recommendation system.In the evaluation part,the performance of MLP in the context-aware module is better than another two supervised models:KNN and SVC.For the results of the whole recommender,the F1-score can arrive at0.55when the length of the recommendation list is set to8.
The contribution of this study includes:(1)Provides an approach for applying the knowledge community and social network analysis theories into the Web services recommendation;(2)Presents a Web services discovering mechanism which can support a service resource platform or an end-user programming application;(3)Presents an approach to apply the resource-allocation in network projection for improving the recommendation performance;(4)Presents a context-aware service recommendation method which can recommend items based on the service composition history,and it is also easy to extended to other contextual information like image and audio.
Mashup,which is a kind of light-weight method for integrating service functions,provides users with a way to integrate existing service resources to meet their needs.As a user-generated content that integrates service resources,mashups show the collaboration relationship between service resources.In the service resource community,this kind of relationship between service resources embodies their functional connection.indicating their functional substitutability and compatibility.Besides,the collaboration network shows the knowledge process within the community,which suggests a way of recommending service according to their functional similarity.However,currently,how to extract this service collaboration network and combine them with service recommendations is still an unsolved question.
The service recommendation method based on context-aware is also a critical way of optimizing the recommendation performance.In response to the complex internet environment,context-aware techniques can mine structured demands expressions from the users situation information and calculate the recommendation results accordingly.
The goal of this study is to fill four gaps:(l)Extract the resources features from service collaboration network metrics and construct a service recommender;(2)Optimize the network projection weights to balance the influence of popular services;(3)Use MLP to extract user demands from textual information;(4)Match the usersdemands with service resources via service collaboration network.
The main work of this study is constructing a service recommendation framework.including two major modules:Recommendation computation based on the service collaboration network extraction and context extraction.The recommendation computation uses network matrix multiplication and resource-allocation process to transform the network into service features.Then the module uses the cosine method to compute the similarity between application functions and services.The context extraction module transforms the text into structured features by using TF-IDF and TSVD.Then the features are used to train a multilayers perceptron,which can learn application context from input features.This module also applies a cross-validation model selector to train the supervised model.
The construction process of the two modules constitutes the offline data processingsubsystem.The users input their demands in natural language via the user interface.and the online recommendation subsystem will use the trained models and recommendation scores to decide the recommendation results.
In the empirical part,this study uses the data from Programmable Web to implement this service recommendation system.In the evaluation part,the performance of MLP in the context-aware module is better than another two supervised models:KNN and SVC.For the results of the whole recommender,the F1-score can arrive at0.55when the length of the recommendation list is set to8.
The contribution of this study includes:(1)Provides an approach for applying the knowledge community and social network analysis theories into the Web services recommendation;(2)Presents a Web services discovering mechanism which can support a service resource platform or an end-user programming application;(3)Presents an approach to apply the resource-allocation in network projection for improving the recommendation performance;(4)Presents a context-aware service recommendation method which can recommend items based on the service composition history,and it is also easy to extended to other contextual information like image and audio.